The objective of this project is to develop method for treating atrial fibrillation (AF) by low- intensity interactive vagal stimulation. AF is the most common arrhythmia in the United States, and has been associated with autonomic modulation of neural substrate. The autonomic nerve activity and its effect on atrial fibrillation has been a subject of significant interests, largely driven by the search for more cost-effective treatments for atrial fibrillation. Evidence from animal studies shows that the initiation of atrial fibrillation is usually accompanied by coactivation of sympathetic and vagal nerves. Therefore, one potential approach to suppress the onset of AF is to inhibit the sympathovagal coactivation before AF occurs. This research will test the hypothesis that low-level vagal stimulation triggered by sympathetic discharges can modify neural substrate of AF and alter AF susceptibility. Such a conceptual AF treatment will be tested in acute and chronic canine models of AF. Unlike the conventional approach of passive vagal stimulation, a new implantable interactive vagal stimulator (I-VS) will be built. The interactive stimulator fires only when the sympathetic activity is high. The aims in this study are 1) Evaluate the efficacy of I-VS in open-chest acute preparations. The I-VS will be designed and tested in acute open-chest preparations with induced AF. The interactive pacing algorithm will detect sympathetic nerve activities for triggering the vagal stimulation. The algorithm of interactive pacing will be first implemented with data acquisition software, followed by hardware design of an implantable neural stimulator based on the system parameters determined from the software operation. The frequency and intensity of the stimulation pulses will be determined based on their ability to alter autonomic discharges during the induced AF. 2) Evaluate the efficacy of I-VS in ambulatory chronic preparations. The interactive neural stimulator will be encapsulated and implanted in animals for later vagal stimulation. The AF will be induced by chronic, intermittent fast pacing of the atrium. The AF inducibility will be compared with and without the interactive vagal stimulation. The I-VS will be again applied during sustained AF to characterize the change of autonomic discharges due to vagal stimulation. The capability to record continuously and interpret in real-time the cardiac nerve activities not only allows characterization of autonomic activities during the progression of heart disease, but also presents as an exciting opportunity for electrical intervention of the autonomic system. Results from this study will prompt new studies to further exploit the potential of autonomic stimulation as a new AF treatment paradigm.